On the complexity of planning for agent teams and its implications for single agent planning

If the complexity of planning for a single agent is described by some function f of the input, how much more difficult is it to plan for a team of n cooperating agents? If these agents are completely independent, we can simply solve n single agent problems, scaling linearly with the number of agents. But if all the agents interact tightly, we really need to solve a single problem that is n times larger, which could be exponentially (in n) harder to solve. Is a more general characterization possible? To formulate this question precisely, we minimally extend the standard STRIPS model to describe multi-agent planning problems. Then, we identify two problem parameters that help us answer our question. The first parameter is independent of the precise task the multi-agent system should plan for, and it captures the structure of the possible direct interactions between the agents via the tree-width of a graph induced by the team. The second parameter is task-dependent, and it captures the minimal number of interactions by the ''most interacting'' agent in the team that is needed to solve the problem. We show that multi-agent planning problems can be solved in time exponential only in these parameters. Thus, when these parameters are bounded, the complexity scales only polynomially in the size of the agent team. These results also have direct implications for the single-agent case: by casting single-agent planning tasks as multi-agent planning tasks, we can devise novel methods for decomposition-based planning for single agents. We analyze one such method, and use the techniques developed to provide some of the strongest tractability results for classical single-agent planning to date.

[1]  Anders Jonsson,et al.  The Complexity of Planning Problems With Simple Causal Graphs , 2008, J. Artif. Intell. Res..

[2]  Adnan Darwiche,et al.  Modeling and Reasoning with Bayesian Networks , 2009 .

[3]  Eyal Amir,et al.  Approximation Algorithms for Treewidth , 2010, Algorithmica.

[4]  David E. Smith,et al.  Planning Under Continuous Time and Resource Uncertainty: A Challenge for AI , 2002, AIPS Workshop on Planning for Temporal Domains.

[5]  Robin Thomas,et al.  Graph Searching and a Min-Max Theorem for Tree-Width , 1993, J. Comb. Theory, Ser. B.

[6]  Craig A. Knoblock Automatically Generating Abstractions for Planning , 1994, Artif. Intell..

[7]  Tom Bylander,et al.  The Computational Complexity of Propositional STRIPS Planning , 1994, Artif. Intell..

[8]  James A. Hendler,et al.  HTN Planning: Complexity and Expressivity , 1994, AAAI.

[9]  R. Halin S-functions for graphs , 1976 .

[10]  Ronen I. Brafman,et al.  Partial-Order Planning with Concurrent Interacting Actions , 2011, J. Artif. Intell. Res..

[11]  Hubie Chen,et al.  Causal graphs and structurally restricted planning , 2010, J. Comput. Syst. Sci..

[12]  Derek G. Corneil,et al.  Complexity of finding embeddings in a k -tree , 1987 .

[13]  Richard Fikes,et al.  STRIPS: A New Approach to the Application of Theorem Proving to Problem Solving , 1971, IJCAI.

[14]  Ronen I. Brafman,et al.  A general, fully distributed multi-agent planning algorithm , 2010, AAMAS.

[15]  Patrik Haslum,et al.  Cost-Optimal Factored Planning: Promises and Pitfalls , 2010, ICAPS.

[16]  Malte Helmert,et al.  Concise finite-domain representations for PDDL planning tasks , 2009, Artif. Intell..

[17]  Bart Selman,et al.  Pushing the Envelope: Planning, Propositional Logic and Stochastic Search , 1996, AAAI/IAAI, Vol. 2.

[18]  Rina Dechter,et al.  Constraint Processing , 1995, Lecture Notes in Computer Science.

[19]  Malte Helmert,et al.  The Fast Downward Planning System , 2006, J. Artif. Intell. Res..

[20]  Eyal Amir,et al.  Factored planning , 2003, IJCAI 2003.

[21]  Ronen I. Brafman,et al.  Factored Planning: How, When, and When Not , 2006, AAAI.

[22]  Carmel Domshlak,et al.  New Islands of Tractability of Cost-Optimal Planning , 2008, J. Artif. Intell. Res..

[23]  Ronen I. Brafman,et al.  Strucutre and Complexitiy in Planning with Unary Operators , 2000, PuK.

[24]  Earl D. Sacerdoti,et al.  Planning in a Hierarchy of Abstraction Spaces , 1974, IJCAI.

[25]  Moshe Tennenholtz,et al.  Multi-entity Models , 1996, Machine Intelligence 14.

[26]  Malte Helmert,et al.  A Planning Heuristic Based on Causal Graph Analysis , 2004, ICAPS.

[27]  Paul D. Seymour,et al.  Graph minors. III. Planar tree-width , 1984, J. Comb. Theory B.

[28]  Avrim Blum,et al.  Fast Planning Through Planning Graph Analysis , 1995, IJCAI.

[29]  Carmel Domshlak,et al.  Implicit Abstraction Heuristics , 2010, J. Artif. Intell. Res..

[30]  Amy L. Lansky,et al.  Scope and Abstraction: Two Criteria for Localized Planning , 1995, IJCAI.

[31]  Hector Geffner,et al.  Branching and pruning: An optimal temporal POCL planner based on constraint programming , 2004, Artif. Intell..

[32]  Bradley J. Clement Abstract Reasoning for Planning and Coordination , 2002, SARA.

[33]  Subbarao Kambhampati,et al.  Planning as constraint satisfaction: Solving the planning graph by compiling it into CSP , 2001, Artif. Intell..

[34]  Makoto Yokoo,et al.  Distributed Constraint Satisfaction , 2000, Springer Series on Agent Technology.

[35]  Ronen I. Brafman,et al.  From One to Many: Planning for Loosely Coupled Multi-Agent Systems , 2008, ICAPS.

[36]  P. Pandurang Nayak,et al.  A Model-Based Approach to Reactive Self-Configuring Systems , 1996, AAAI/IAAI, Vol. 2.

[37]  Ronen I. Brafman,et al.  Transferable Utility Planning Games , 2010, AAAI.

[38]  Georg Gottlob,et al.  Fixed-parameter complexity in AI and nonmonotonic reasoning , 1999, Artif. Intell..

[39]  Malte Helmert,et al.  Complexity results for standard benchmark domains in planning , 2003, Artif. Intell..